JP5684413B2 - Catadioptric projection objective with pupil mirror, projection exposure apparatus and method - Google Patents

Description

  The present invention relates to a catadioptric projection objective for imaging an object field arranged in the object plane of a projection objective onto an image field arranged in the image plane of the projection objective.

  Catadioptric projection objectives are employed in, for example, projection exposure systems used to fabricate semiconductor devices and other types of microdevices, specifically wafer scanners or wafer steppers, and are generally referred to below as “masks”. Alternatively, it achieves the role of projecting a photomask or pattern on the reticle, referred to as a “reticle”, onto an object having a photosensitive coating, using ultra-high resolution on a reduced scale.

  In order to create even finer structures, both seeking to increase the image-side numerical aperture (NA) of projection objectives and adopting ultraviolet light with shorter wavelengths, preferably shorter than about 260 nm, was explored. Has been. However, there are only a few materials that are sufficiently transparent in this wavelength range available to fabricate optical elements, specifically synthetic quartz glass and fluoride crystals. The Abbe numbers of these materials available are quite close to each other and it is difficult to achieve a purely refractive system that is sufficiently well color corrected (corrected for chromatic aberration).

  In optical lithography, high resolution and good correction status must be obtained in a relatively large, substantially flat image field. It has been pointed out that the most difficult requirement that can be sought for any optical design is to have a flat image, especially for all refractive designs. Achieving a flat image requires an opposing lens power, which results in a larger and higher lens resulting from a stronger lens, longer system length, larger system glass mass, and stronger lens curvature. The following image aberrations are caused.

Concave mirrors have been used for some time to help solve the problems of color correction and image flattening. The concave mirror has a positive refractive power like a positive lens, but does not have an opposite sign of Petzval curvature. Concave mirrors do not cause color problems. Accordingly, catadioptric systems that combine refractive and reflective elements, in particular lenses and at least one concave mirror, are mainly employed in constructing high resolution projection objectives of the type described above.
Unfortunately, concave mirrors are difficult to integrate into an optical design because they send the emitted light back in the direction of arrival. An intelligent design that integrates concave mirrors without causing mechanical problems or problems due to beam vignetting or pupil obscuration is desirable.

  Increasing demands on the efficiency of the lithographic manufacturing process tend to increase the power of the light source. Shorter wavelengths are also being used. In order to optimize the imaging conditions in various types of patterns, specific lighting environments are employed. As a result, various time dependent changes in the properties of optical materials in the projection system that significantly affect the imaging quality of the exposure system have been observed. Heating of the active lens group and other transmissive optical elements due to high absorption (lens heating) is one effect that dynamically affects the imaging properties. In addition, long-term (pseudo-static) effects such as consolidation effects due to changes in refractive index induced by synchrotron radiation have been observed.

  Applicant's patent application US 2004/0144915 A1 reveals a technique for solving some problems caused by the heating effect induced by absorption in a catadioptric projection objective having a physical beam splitter. This application discloses a folded catadioptric projection objective designed as a single imaging system (without an intermediate image) with a concave mirror positioned on the pupil surface. A physical beam splitter having a polarization-selective beam splitter surface is provided to separate the radiation coming from the on-axis object field towards the concave mirror from the radiation reflected by the concave mirror towards the image plane. The concave mirror is configured as a deformable mirror, and the shape of the concave mirror surface cancels certain time-dependent imaging aberrations that progress during the operation of the projection objective in response to changes in optical performance induced by the emitted light It can be operated using a pupil mirror manipulator in such a way that it is possible to do so. The pupil mirror manipulator has a simple structure and is provided on the back side of the concave mirror so as not to interfere with the optical path. Deformable pupil mirrors are designed to compensate for, for example, astigmatism, double or quadruple wavefront deformation, and consolidation effects due to absorption-induced heating of the cubic beam splitter and rectangular retardation plate. The

US2004 / 0144915A1 WO2004 / 019128A2 WO2005 / 111689A US6, 995, 833B2 US6, 600, 608 WO2005 / 098506A1 WO2005 / 098505A1 US patent application Ser. No. 11 / 151,465 US2006 / 0039669 WO2003 / 093903 US6,252,647B1 US2006 / 005026A1 EP1069448B1 US2002 / 0001088A1 US6,784,977B2 US 5,986,795 US2006 / 0018045A1 US2002 / 0011088A1 US2006 / 0114437A1 WO2005 / 040890A

The object of the present invention is for microlithography suitable for use in the vacuum ultraviolet (VUV) range, which has the potential for very high image-side numerical aperture and can maintain long-term stability of performance during operation. A catadioptric projection objective is provided.
Another object of the present invention is to provide a projection objective and an exposure apparatus configured to operate at stable operating conditions in a variety of illumination environments including non-conventional off-axis illumination such as dipole and quadrupole illumination. It is to be.

Another object of the present invention is to provide a catadioptric projection objective that provides stable optical performance in immersion lithography.
Another object of the present invention is for microlithography applications at wavelengths up to 193 nm that can be manufactured at reasonable cost and have optical performance stability under various illumination conditions of dry or immersion lithography. A catadioptric projection objective is provided.

  As a solution to these and other objects, the present invention, according to one configuration, images a pattern from an object field located in the object plane of the projection objective onto an image field located in the image plane of the projection objective. A catadioptric projection objective for imaging a pattern from an object plane into a first intermediate image, a first objective part having a first pupil surface; A second objective part having a second pupil surface optically conjugate with the first pupil surface, the second objective part being configured to form one intermediate image in the second intermediate image; A third objective portion configured to form an image in the image plane and having a third pupil surface optically conjugate with the first and second pupil surfaces; A pupil mirror having a reflective pupil mirror surface positioned at or near one of the third pupil surfaces; It is operatively connected to the mirror, and a pupil mirror manipulator configured to change the shape of the pupil mirror surface.

  Changes in the properties of optical materials (lens heating, consolidation, etc.) in the various time-dependent radiation-induced projection objectives presented at the beginning of this application induce unique imaging aberrations and these imaging Aberration is difficult to compensate with a conventional manipulator because the main contribution to imaging aberration is the constant field of view contribution to aspherical aberration. More specifically, constant axial astigmatism (AIDA) and quadruple radial symmetry error (four undulations) are observed over the field of view. By providing a pupil mirror having a pupil mirror surface and an operating means that enables the shape of the reflective pupil mirror surface to be changed, the wavefront shape is significantly different in the projection objective due to lens heating and compaction. A wavefront that is essentially free of aberrations in the image plane is obtained. In general, correction can be obtained by adjusting the shape of the reflective pupil mirror surface so that the negative contribution of the disturbance caused by lens heating or the like is offset by the pupil mirror. Since the pupil mirror is positioned at or near the pupil surface of the projection objective, any deformation or modification of the reflecting surface shape has essentially the same effect at all field points, thereby making the field of view essentially constant. Correction is obtained.

  A catadioptric projection objective with two real intermediate images has a very high image side numerical aperture in the image field that is large enough to allow microlithographic applications while avoiding problems such as vignetting. It has been found that it can be designed to obtain. Further, in cases where off-axis object field and image field are used, pupil obscuration can be avoided in systems with high image side NA. This projection objective has exactly three consecutive objective parts and exactly two intermediate images. Each of the first to third objective parts is an imaging subsystem (2f system) that performs two successive Fourier transforms, and in addition to the first to third objective parts, additional There is no special objective part. If exactly two intermediate images are provided, a high degree of freedom for the optical designer is achieved in an optical system that can be produced with reasonable size and complexity. A large image side aperture in the image field suitable for lithographic purposes is allowed.

Although the pupil mirror can have an essentially flat reflective surface, in most embodiments the pupil mirror is designed as an optical element with refractive power. In some embodiments, the pupil mirror is a concave mirror.
It has proved useful to form at least one intermediate image between the object plane and the pupil mirror, so that the emitted light can be properly prepared before reflection on the pupil mirror. Become. If at least one imaging subsystem (objective part) is provided between the object plane and the pupil mirror, the emitted light impinging on the pupil mirror is substantially prepared for optimal use of the pupil mirror correction function. can do.

In some embodiments, the pupil mirror is located in or near the second pupil surface within the second objective section. In order to form a catadioptric second objective part, one or more lenses can additionally be provided in the second objective part. Alternatively, the second objective part can be purely reflective (reflective imaging).
In some embodiments, the first objective portion that forms the first intermediate image is purely refractive, i.e. includes only one or more lenses and does not include an imaging mirror. Alternatively or additionally, the third objective part that forms the final image in the image plane from the second intermediate image may be a purely refractive objective part.

The following comprises a first refractive objective part (R), a second catadioptric or reflective imaging objective part (C) and a refractive (R) third objective part in the order described above. The connected system will be represented by an “R—C—R” type system.
In some embodiments, the first objective portion is a catadioptric objective portion, the second objective portion is of catadioptric or reflective imaging, includes a pupil mirror, The objective instrument part is a refractive objective instrument part. These systems can be referred to as “CCR” type systems.

The optical elements of the projection objective can be arranged in various ways.
Some embodiments are designed as “folded” catadioptric projection objectives with optical axes that are subdivided into at least two non-parallel axis segments using mirrors (deflection mirrors). In general, the deflecting mirror can have a flat reflecting surface, that is, it can have no refractive power.
Some embodiments deflect the emitted light from the object plane toward the pupil mirror or pupil so that a geometrically formed double pass region is formed between the first deflecting mirror and the pupil mirror. It arrange | positions so that radiation light may be deflected toward an image surface from a mirror. Providing at least one deflecting mirror facilitates placing the pupil mirror at or near the pupil surface without unduly limiting the available numerical aperture by the size of the pupil mirror.

A second deflection mirror can be provided that is disposed at an angle of 90 ° with respect to the first deflection mirror so that the object plane and the image plane are parallel. The second deflection mirror can be arranged to directly receive the radiation reflected from the pupil mirror, or can be arranged to receive the radiation reflected from the first deflection mirror. it can.
In some embodiments, the first deflecting mirror is arranged to deflect radiation coming from the object plane in the direction of the pupil mirror, and the second folding mirror images radiation coming from the pupil mirror. Arranged to deflect in the direction of the plane. This folding geometry is essentially coaxial with the optical axis segment defined by the optical element of the first objective part and the optical axis segment defined by the optical element of the third objective part, i.e. strictly It can be arranged with a slight lateral offset which is coaxial to the lens or very small in relation to the general lens diameter. Examples of projection objectives with this general folding geometry are disclosed in WO2004 / 019128A2 or WO2005 / 111689A. The disclosures of these documents are incorporated herein by reference.

  The first deflecting mirror is optically disposed downstream of the pupil mirror so as to deflect the radiation reflected by the pupil mirror toward the second deflecting mirror, and the second deflecting mirror is the first deflecting mirror. It arrange | positions so that the emitted light from a mirror may be deflected toward an image surface. In these types of embodiments, the optical axis defined by the concave pupil mirror can be coaxial with the segment of the optical axis defined by the first objective part. Generally, a large lateral offset is obtained between the object plane and the image plane, and this offset is large compared to a typical lens diameter. Generally this type of objective includes two lens barrel structures mounted parallel to each other. General examples are disclosed, for example, in US 6,995,833B2. The disclosures of these documents are incorporated herein by reference.

  A negative group comprising at least one negative lens can be arranged in a double-pass region where the emitted light in front of the reflecting side of the concave mirror passes through the negative group in the opposite direction at least twice. . This negative group can be positioned in the immediate vicinity of the concave pupil mirror in a region near the pupil, which has an imaging peripheral ray height (MRH) greater than the chief ray height (CRH). It can be characterized by the fact that Preferably, the marginal ray height is at least twice as large as the principal ray height in the negative group region, in particular at least 5 to 10 times greater. Negative groups in the region of large marginal ray height can contribute substantially to color correction, especially the axial chromatic aberration of thin lenses is proportional to the square of marginal ray height at the lens location. (Further, it is proportional to the refractive power and the dispersion of the lens), so that it can substantially contribute to the correction of axial chromatic aberration. To this is added the fact that the projected radiation passes twice in the opposite radiation passage direction through the negative group placed in the immediate vicinity of the concave mirror, resulting in two negative overcorrection effects of the negative group color. Used. The negative group can comprise, for example, a single negative lens, or can include at least two negative lenses.

In some embodiments, all optical elements of the projection objective are aligned along a straight optical axis that is common to all optical elements of the projection objective. This type of optical system is referred to herein as a “linear arrangement system”.
From an optical point of view, a linear arrangement system is preferred because optical problems caused by utilizing a flat folding mirror such as the polarization effect can be avoided. Also from a manufacturing point of view, the linear placement system can be designed to take advantage of conventional mounting techniques for optical elements, thereby improving the mechanical stability of the projection objective. By utilizing an even number of mirrors, such as two or four or six mirrors, imaging without image inversion is possible.

The optical element of the linear arrangement system includes an object-side mirror group entrance for receiving radiation light from the object plane, and an image-side mirror group exit for emitting radiation light that emerges from the mirror group exit toward the image plane. The mirror group includes at least one pupil mirror as described above.
If an on-axis field (the object field and the image field are centered with respect to the optical axis) is desired, the mirror group can be formed by a pair of concave mirrors whose concave reflecting surfaces face each other and the emitted light is mirrored. In order to be able to pass through, a transmissive part (such as a hole or hole) is made in the area around the optical axis in the mirror surface. The concave mirror can be placed optically close to the pupil surface. At least one of the concave mirrors may be provided with a pupil mirror manipulator to form a deformable pupil mirror. An example of a system with two intermediate images, an on-axis field of view, and a pupil obscuration is disclosed, for example, in Applicant's patent US 6,600,608, the disclosure of which is incorporated herein by reference. ing.
If imaging without pupil obscuration is desired, an off-axis field (the object field and the image field are completely outside the optical axis) can be used.

  In general, the linear arrangement system has a very small installation space for positioning the mirror. In addition, the mirror size on the pupil surface limits the function of forming an “effective image field” of an appropriately sized rectangular or arc shape with a large image-side numerical aperture. This is a relatively low value of “effective etendue” (effective geometric beam) of the object field, ie between the innermost field point that can be imaged without vignetting and the outer edge of the effective object field. This corresponds to a small distance. Under these conditions, when a moderately sized effective object field having a rectangular or arcuate shape is desired, the "design object field", i.e. the field size that the projection objective should be adequately corrected is Become bigger. When trying to increase the size of the design object field, the number and size of optical elements generally increases rapidly, so it is generally desirable to keep the design object field as small as possible. For a detailed definition of the terms “object field”, “effective etendue” and the interrelationship between these terms, the applicant's international patent application WO 2005 /, whose disclosure is incorporated herein by reference. 098506A1).

  For at least these considerations, the mirror group includes a first mirror for receiving radiation from the mirror group entrance on the first reflection area, and radiation reflected from the first mirror. A second mirror for receiving light on the second reflection area, a third mirror for receiving radiation light reflected from the second mirror on the third reflection area, and reflection from the third mirror A third mirror for receiving the emitted radiated light and reflecting the radiated light to the mirror group exit, wherein at least two of the mirrors are concave mirrors having a rotationally symmetric curvature surface with respect to the optical axis It has proved useful.

By providing at least four mirrors in the mirror group, it is possible to limit the size of the pupil mirror even when the image-side numerical aperture increases, thereby facilitating vignetting control. Preferably exactly four mirrors are provided. All mirrors in the mirror group can be concave mirrors.
Although it is possible to use the second mirror (which is geometrically close to the object plane) as a pupil mirror, it is often useful to configure the third mirror as a pupil mirror. It turns out. In general, the third mirror is geometrically away from the object plane and is on the image side of the mirror group, thereby providing a geometrical shape for appropriately directing the emitted light towards the third mirror. Space is given. Also, an optical element that is optically upstream of the third mirror and that includes the first and second mirrors can be utilized to properly shape and prepare the emitted light beam towards the pupil mirror. . For example, the correction status and chief ray height can be changed as needed.

  Preferably, the mirrors of the mirror group are formed so that the radiated light coming from the mirror group entrance crosses the optical axis before exiting the mirror group at the mirror group exit, and geometrically shaped to the mirror group entrance and the mirror group It is arranged to pass at least five times through the mirror group plane arranged between the outlets. As a result, at least four reflections can be obtained in an axially compact space defined between the mirror group entrance and the mirror group exit.

  The front lens group can be placed between the object plane and the mirror group entrance, thereby converting the spatial distribution of the emitted light at the object plane into the desired angular distribution of the emitted light at the mirror group entrance, Is incident on the mirror group, and the incident angle on the first mirror can be adjusted. The design of the front lens group allows the emitted light beam incident on the mirror group entrance to have a desired cross-sectional shape, allowing the emitted light beam to pass through the mirror group entrance without hitting the adjacent mirror edge, It can be selected so that beam vignetting is avoided. The front lens group is a Fourier lens group, i.e. a single optical element that performs one single Fourier transform or an odd number of continuous Fourier transforms between the front and back focal planes of the Fourier lens group, or It can be designed as a group of at least two optical elements. In a preferred embodiment, the Fourier lens group forming the front lens group is purely refractive and performs a single Fourier transform. In a preferred embodiment, the Fourier lens group is configured to image the entrance pupil of the projection objective essentially at the position of the mirror group entrance such that there is a pupil surface at or near the mirror group entrance. . Embodiments without a front lens group are also possible.

A catadioptric linear arrangement system having a compact mirror group of four mirrors is disclosed in the applicant's international patent application WO2005 / 098505A1, the disclosure of which is incorporated herein by reference. Some embodiments include a pupil mirror and can be used in connection with the present invention with appropriate modifications.
In some embodiments, the first objective part (forming an intermediate image upstream of the pupil mirror) is designed as a magnified imaging system with a magnification ratio | β |> 1, thereby reducing the effective object field. Is also formed. Preferably, the condition | β |> 1.5 holds. The magnified intermediate image can be used to obtain a large principal ray angle CRA _PM at the downstream pupil mirror. Considering that the product of paraxial principal ray angle CRA and pupil size is constant in an optical imaging system (Lagrange invariant), a large chief ray angle at the pupil surface is a small pupil, ie at the pupil surface. Corresponds to the small beam diameter of the beam.

In some embodiments, the pupil mirror is optically disposed between a first mirror upstream of the pupil mirror and a second mirror downstream of the pupil mirror, wherein the chief ray height is the object plane a CRH _O at an inner, a CRH ₁ in the first mirror, a CRH ₂ in the second mirror, conditions CRH _1> CRH _O and CRH _2> CRH _O is satisfied. Preferably, at least one of the conditions CRH ₁ > 1.5 × CRH _{2 O} and CRH ₂ > CRH _{2 O} is satisfied. In other words, the ray height of the principal ray in the mirror immediately upstream and immediately downstream of the pupil mirror is larger than the object height. A chief ray height CRH _O in the object plane, the ratio between the chief ray height at least one of the upstream or downstream of the mirror of the pupil mirror optically, for example, at least 1.5, or at least 2.0, Or at least 2.5. Under these conditions, a small beam diameter is obtained in the pupil mirror, which makes it possible to have a small size pupil mirror.

In some embodiments, the optical element disposed between the object plane and the pupil mirror is configured to achieve a maximum chief ray angle CRA _max > 25 ° in the pupil mirror. In some embodiments, a maximum chief ray angle greater than 30 °, or greater than 35, or even greater than 40 ° is possible. A small pupil allows the use of a pupil mirror with a small diameter. Furthermore, it makes it possible to guide the radiated light through the pupil mirror even when the radiated light has a large aperture in the projection beam. Therefore, a large chief ray angle at the pupil mirror makes it easy to obtain a high image side numerical aperture in a catadioptric linear arrangement system.

The optically utilized free diameter _DPM of the pupil mirror can be extremely small. In some embodiments, the pupil mirror is an optical element that has the smallest diameter in the overall projection objective. The pupil mirror diameter D _PM may be less than 50%, or less than 40%, or less than 30% of the maximum free diameter of the optical element in the projection objective.
The projection objective can have an aperture stop that allows the aperture diameter to be adjusted as required, with the maximum aperture diameter of the aperture stop being at least twice as large as the pupil mirror diameter D _PM. is there.

If the deformation of the pupil mirror surface is assumed to produce a correction that is essentially the same at all field points (herein referred to as a “constant field of view” correction), the pupil Special emphasis should be given to the correction status of the projection beam incident on the mirror. Preferably, the projection beam at the pupil mirror complies with the condition:
| CRH _i | / D ₀ <0.1 (1)
0.9 ≦ D _i / D ₀ ≦ 1.1 (2)
Where | CRH _i | is the principal ray height quantity of the principal ray at the object field point i on the pupil mirror surface, and D ₀ is twice the marginal ray height quantity on the pupil mirror surface. Yes, if HRR _i is the marginal ray height of the upper marginal ray corresponding to field point i and HRRL _i is the marginal ray height of the lower marginal ray corresponding to field point i, then D _i = | HRRU _i -HRRL _i | is the meridional diameter at the pupil mirror surface of the image of the entrance pupil of the projection objective for field point i.
The chief ray is a ray from the outermost field point with respect to the center of the entrance pupil. If the above conditions (1) and (2) are met, the image of the entrance pupil is positioned at or very close to the pupil mirror, so that a constant field of view is corrected according to the deformation of the pupil mirror. It becomes possible to obtain.

  In some embodiments, if a compact design in the axial direction is desired, a pupil mirror and another mirror with their respective reflective surfaces of the same general orientation should be placed geometrically close to each other May be necessary. In some embodiments, this problem is solved by providing a mirror pair composed of two concave mirrors having a mirror surface sharing a common curvature surface provided on a common substrate, one of the concave mirrors. Is a pupil mirror having a reflective pupil mirror surface configured to be deformable by a pupil mirror manipulator, the other concave mirror having a rigid refractive surface separate from the pupil mirror surface. This mirror combination can be provided in one common mounting structure, which facilitates the mounting of the pupil mirror.

  In some embodiments, the pupil mirror is disposed at or near one of the pupil surfaces, and one or more transmissive optical elements are at least one of the optically conjugated pupil surfaces. One or close to it. In the present description, in general, the position “at or near” the pupil surface is such that the peripheral ray height MRH is the principal ray height CRH such that the ray height ratio is RHR = MRH / CRH> 1. Can be characterized as a larger position. Absorption-induced deformations and refractive index changes induced in these lens elements (close to the conjugate pupil surface) can cause wavefront aberrations, and these wavefront aberrations appropriately deform the pupil mirror surface It can be compensated by the targeting method. For example, when a polar illumination environment such as dipole illumination or quadrupole illumination is used, a non-uniform radiant light load having double or quadruple radiation symmetry, respectively, is transmitted through the transmissive optical element proximate the pupil surface Can occur within. The resulting wavefront deformation, which may have double or quadruple radial symmetry, corresponds to the pupil mirror surface by essentially deforming the pupil mirror surface with double or quadruple radial symmetry. The deformation can be at least partially compensated.

The compensation function brought about by the operation in the pupil mirror can be particularly sensitive to heating effects induced by non-uniform (non-uniform) absorption, depending on the properties of the optical material used. It can be particularly useful in the system. For example, it may be desirable to use fused silica (synthetic quartz glass) to manufacture some or all of the projection objective lenses. Fused silica is available in large quantities and sizes to produce high optical quality and large lenses as required by high NA microlithographic projection objectives. Furthermore, there is a sufficient track record for treating fused silica to obtain high quality optical surfaces. Further, fused silica has virtually no absorption until it has a wavelength as short as about 190 nm. Thus, it may be desirable to use a large amount of fused silica towards the objective lens. On the other hand, the specific heat conductivity of fused silica is smaller than that of calcium fluoride (CaF ₂ ) and other alkali fluoride crystal materials, and these materials absorb even at short wavelengths such as 157 nm. It is very low and is used for these short wavelengths. The specific heat conductivity of calcium fluoride is greater than that of fused silica, and local temperature gradients caused by non-uniform heating can average faster and more efficiently in materials with relatively high specific heat conductivity. So, the negative effect of non-uniform heating by projection radiation can be less with calcium fluoride compared to fused silica, so that the system made with the above materials can produce non-uniform lens heating Less susceptible to problems caused by. By providing a pupil mirror that can be manipulated to compensate for lens heating effects, for example, under certain lighting conditions (eg, dipole or quadrupole illumination) and / or being imaged With the use of a constant pattern structure, it is possible to use fused silica even at or near the conjugate pupil surface where the non-uniform heating of the lens material is likely to occur.

In some embodiments, the optical element at or near the optically conjugated pupil surface is made of an optical material having a specific heat conductivity that is less than the specific heat conductivity of calcium fluoride. The optical elements at or near the optically conjugated pupil surface can be made, for example, from fused silica.
In some embodiments, at least 90% of all lenses of the projection objective are made from fused silica. In some embodiments, all lenses are made from fused silica.

  Embodiments can have, for example, an image-side numerical aperture NA ≧ 0.6, which allows for use in microlithography and results in a small feature size in a microlithographic exposure process. Embodiments with NA ≧ 0.7 are possible. In some embodiments, the catadioptric projection objective is a “dry system”, ie, an image space (where the substrate is located) between the exit surface of the projection objective and the image plane where the refractive index is close to unity. Image-side numerical aperture NA ≧ 0.8 or even NA ≧ 0.9 close to the theoretical limit in a projection objective adapted for imaging aberrations in a dry process filled with a gas having

In another embodiment, the catadioptric projection objective is imaged in a wet process where the image space between the exit surface and the image plane of the projection objective is filled with an immersion medium having a refractive index significantly greater than 1. Designed as an immersion objective adapted for aberrations. For example, the refractive index can be 1.3 or greater, or 1.4 or greater, or 1.5 or greater. The projection objective, for example, when used with an immersion medium having a refractive index n _l > 1.3, image side numerical aperture NA> 1.0, eg NA ≧ 1.1, or NA ≧ 1.2, Or NA ≧ 1.3. Alternatively, the projection objective can have an image side numerical aperture NA <1.0 when used with an immersion medium.

In general, the image-side numerical aperture NA is limited by the refractive index of the surrounding medium in the image space. In immersion lithography, the theoretically possible numerical aperture NA is limited by the refractive index of the immersion medium. The immersion medium can be liquid (liquid immersion, “wet processing”) or solid (solid immersion).
For practical reasons, the aperture should not be arbitrarily close to the refractive index of the last medium, i.e. the medium closest to the image, which results in a very large propagation angle relative to the optical axis. Because. Empirically, the image side NA can be close to about 95% of the refractive index of the last medium on the image side. In immersion lithography at λ = 193 nm, this corresponds to a numerical aperture NA = 1.35 with water (n _H2O = 1.43) as the immersion medium.

  Some embodiments are configured to allow the range of the image side NA to be extended beyond the value NA = 1.35 and beyond. In some embodiments, the at least one optical element is a high refractive index optical element made from a high refractive index material having a refractive index n ≧ 1.6 at the operating wavelength. The refractive index may be about 1.7 or greater, or 1.8 or greater, or even 1.9 or greater at the operating wavelength. The operating wavelength can be in the deep ultraviolet (DUV) region shorter than 260 nm, such as 248 nm or 193 nm.

  The projection objective has a final optical element that is closest to the illumination system. The exit side of the final optical element in the immediate vicinity of the illumination system forms the exit surface of the projection objective. The exit surface can be flat or curved, eg concave. In some embodiments, the final optical element is at least partially made of a high refractive index material having a refractive index n> 1.6 at the operating wavelength. For example, the final optical element can be an integral plano-convex lens with a spherical or aspheric curvature entrance surface and a flat exit surface proximate to the illumination system.

The high refractive index material can be, for example, sapphire (Al ₂ O ₃ ), which can be used as a high refractive index material until it is as short as about λ = 193 nm. In some embodiments, the high refractive index material is lutetium aluminum garnet (LuAG) having a refractive index n = 2.14 at λ = 193 nm. The high refractive index material can be barium fluoride (BaF ₂ ), lithium fluoride (LiF), or barium lithium fluoride (BaLiF ₃ ). An immersion liquid having a higher refractive index than that of pure water, such as water with an additive for increasing the refractive index, or cyclohexane having n _l = 1.65 at 193 nm can be used. In some embodiments, at λ = 193 nm, image-side numerical apertures NA> 1.4 and NA> 1.5, such as NA = 1.55, can be obtained. The use of high refractive index materials in catadioptric projection objectives for immersion lithography is disclosed, for example, in commonly assigned US patent application Ser. No. 11 / 151,465. The disclosure of this document regarding the use of high index materials is incorporated herein by reference.

The present invention makes it possible to provide a medium size catadioptric projection objective that allows immersion lithography, particularly at image side numerical aperture NA> 1. The imaging characteristics of the projection objective can be modified dynamically by adjusting the shape of the reflective pupil mirror surface during the operation time of the projection objective. In the pupil mirror, if certain conditions regarding the correction status of the projection beam are observed, an essentially constant (slight change or no change) correction effect over the field of view (constant field of view correction) can be obtained. . In an immersion system, these functions can be used, for example, to compensate for imaging aberrations caused by time-dependent changes in the optical properties of the immersion liquid that can be caused by temperature changes during operation. For example, a constant field of view contribution to spherical aberration caused by immersion layer temperature drift can be compensated for in a substantially telecentric image at the wafer.
This compensation can be used independently of the type of projection objective (eg an immersion or folding projection objective with or without an intermediate image).

  According to another aspect, the present invention also relates to a method of fabricating a semiconductor device and other types of microdevices utilizing a catadioptric projection objective, which includes a mask that provides a defined pattern of the projection objective. Locating in the object plane, illuminating the mask with ultraviolet light having a specified wavelength, and a substrate having an exit surface of the projection objective and a substrate surface disposed substantially in the image plane of the projection objective In between, disposing an immersion layer formed by an immersion liquid having a refractive index substantially greater than 1, projecting an image of the pattern onto the photosensitive substrate through the immersion layer, and a projection objective Adjusting the imaging characteristics of the projection objective by changing the surface shape of the pupil mirror having a reflective pupil mirror surface positioned at or near the pupil surface of the projection objective.

The adjustment step can be performed during exposure of the substrate at the location of use of the projection objective and / or during exchange of the substrate and / or between different masks and / or during changes between different illumination environments. .
It has been observed that the lens heating effect may be particularly clear when an illumination environment is used that has only a small degree of interference, such as dipole or quadrupole illumination. Under these conditions, transmissive optical elements such as lenses positioned at or near the pupil surface of the projection objective can be exposed to spatially non-uniform radiation loads and are locally Local heating in the region of maximum peak of the emitted light intensity and maximum diffraction of the imaging is caused. This can lead to inherent deformations and refractive index changes of lens elements having approximately double or quadruple radial symmetry, for example depending on whether dipole or quadrupole illumination is used. Furthermore, these deformations and refractive index changes can cause corresponding wavefront deformations, degrading imaging performance.

  A solution to these and other problems is the use of catadioptric projection objectives to fabricate semiconductor elements and other types of microdevices, which include masks that provide a defined pattern in projection objectives. Locating in the object plane; illuminating the mask with ultraviolet light having a defined wavelength utilizing an illumination environment defined by the illumination system; and in the pupil surface of the illumination system and at least one of the projection objectives Adjusting the illumination within the optically conjugated pupil surface to form an off-axis illumination environment in which the light intensity in the region outside the optical axis is greater than or near the optical axis; and the projection objective Off-axis illumination so that wavefront aberrations caused by spatially non-uniform radiation loading on optical elements at or near the pupil surface of the instrument are at least partially compensated Adjusting the imaging characteristics of the projection objective by changing the surface shape of the pupil mirror having a reflective pupil mirror surface positioned at or near the pupil surface of the projection objective in a manner adapted to the environment; including.

  In general, the lighting environment can be set according to the type of pattern provided by the mask or another patterning means. The lighting environment can be changed when a first reticle having a predetermined structure is replaced by a second reticle having another structure. In an exposure method using multiple exposures, two or more different illumination environments, optionally including off-axis illumination environments, can be used to illuminate a given structure in successive exposure steps. The off-axis illumination environment can be a polar illumination environment such as dipole illumination or quadrupole illumination.

  Since some of the aberrations caused by changes in the properties of the immersion layer and / or projection objective can be dynamic effects on a relatively short time scale, optionally the optical properties of the immersion layer Imaging aberrations related to time-dependent changes in the optical properties of the projection system, including time-dependent changes and / or changes induced by the off-axis pole illumination environment being utilized, etc. (or other characteristics that affect imaging aberrations) ) Are detected (or sensed) by an appropriate detector (or sensor) to generate a sensing signal indicative of these time-dependent changes, and further deformation of the pupil mirror surface shape at least partially compensates for the time-dependent changes. Driving the shape change of the pupil mirror surface by the pupil mirror manipulator in response to the sensing signal can be achieved so as to be effective to do so. Thereby, the pupil mirror manipulator is integrated into the control loop, allowing real-time control of imaging aberrations. In particular, the pupil mirror manipulator can be driven such that the constant field aberration contribution caused by the refractive index change of the immersion liquid is at least partially compensated. Thereby, a more stable process of immersion lithography can be obtained.

  For example, imaging aberrations can be detected directly using an interferometer or other suitable direct measurement system. An indirect method is also possible. For example, if an immersion liquid is used, a temperature sensor can be provided to monitor the temperature of the immersion liquid forming the immersion layer, and the refractive index change caused by the temperature change can be measured in these measurements. It can be empirically derived from the value based on a lookup table and compensated by dynamically manipulating the pupil mirror.

Alternatively or additionally, feedforward control of operation can be utilized to obtain the desired pupil mirror shape. For example, the control unit can receive a signal indicating the type of polar illumination environment set in the illumination system, induced by locally non-uniform absorption in the lens at or near the pupil surface of the projection objective. Appropriate control signals can be provided to the pupil mirror manipulator, such as based on a look-up table, to deform the pupil mirror surface so that wavefront deformation caused by the applied heating is at least partially compensated.
These and other characteristics are not limited to the claims, but individual features may be used individually or in partial combinations as individual or partial combinations of embodiments of the present invention and in other fields. It can also be seen in the description and drawings that can be represented in

1 is a lens cross-sectional view of a linearly arranged catadioptric dry objective for microlithography of the first embodiment having a deformable concave pupil mirror with NA = 0.98. FIG. FIG. 2 is a schematic axial section through a catadioptric linear arrangement projection objective similar to the embodiment of FIG. 1; FIG. 2 is an enlarged detail view of a region around a mirror group in the embodiment of FIG. 1 including a pair of concave mirrors on a common substrate forming an adaptive pupil mirror and a mirror illuminated off-axis. It is the schematic which shows the characteristic which influences the correction status of the projection beam in a pupil mirror. FIG. 6 is a lens cross section of a catadioptric linearly arranged projection objective for immersion lithography of a second embodiment using a rectangular effective object field with NA = 1.2. FIG. 9 is a lens cross-sectional view of a catadioptric linear array projection objective adapted for immersion lithography of a third embodiment using NA = 1.55 and an arcuate effective object field (annular field). FIG. 6 is a schematic axial cross-section through another embodiment of a catadioptric linear arrangement projection objective having a group of four concave mirrors including a pupil mirror. FIG. 6 is a schematic axial cross-section through another embodiment of a catadioptric linear array projection objective having a mirror group of four concave mirrors including two pupil mirrors. 1 is a schematic view of a scanning projection exposure system for microlithography with a catadioptric projection objective having four illumination mirrors including an illumination system and a deformable pupil mirror designed to create a slit-shaped illumination field; FIG. FIG. 6 is a lens cross-sectional view of a catadioptric linear array projection objective adapted for dry lithography with NA = 0.75 and an arcuate effective object field (ring field). FIG. 6 is a lens cross section of a folded catadioptric projection objective adapted for immersion lithography of an embodiment using a rectangular effective object field with NA = 1.25.

In the following description of the preferred embodiment, the object involved is a mask (reticle) that holds a pattern of an integrated circuit layer or any other pattern, for example a diffraction grating pattern. The object image is projected onto a wafer that serves as a substrate coated with a photoresist layer, but other types of substrates such as liquid crystal display components or substrates for optical diffraction gratings are used as well. Is possible.
An embodiment having a plurality of mirrors will be described. Unless otherwise explained, the mirrors are numbered according to the order in which the emitted light is reflected on the mirrors. In other words, the mirror number represents the mirror according to its position along the optical path of the emitted light, not according to its geometric position.
Where appropriate, equal or similar features or groups of features in different embodiments are represented with similar reference identifications.
Where tables are provided to disclose the design specifications shown in the figures, single or multiple tables are indicated by the same numbers as the respective figures.

In some embodiments described below, the curvature surfaces of all curvature mirrors have a common axis of rotational symmetry, also referred to as the mirror group axis. The mirror group axis coincides with the optical axis OA of the projection objective. An axisymmetric optical system, also referred to as a coaxial system or a linear arrangement system, is provided in this manner. The object plane and the image plane are parallel. An even number of reflections occur. Effectively used object and image fields are off-axis, i.e., completely positioned outside the optical axis. All systems have a circular pupil centered on the optical axis, which allows it to be used as a projection objective for microlithography.
In other embodiments, the optical axes are folded into axial segments that are inclined at an angle relative to each other.

  FIG. 1 shows that a pattern image on a reticle arranged in a flat object plane OS (objective plane) is reduced on a flat image plane IS (image plane) while creating two actual intermediate images IMI1 and IMI2. 1 shows a lens cross section of a first embodiment catadioptric projection objective 100 designed to project at a scale, eg, 4: 1. Thereby, the off-axis effective object field OF positioned outside the optical axis OA is projected onto the off-axis image field IF. FIG. 2 shows a schematic view of a version of a projection objective of the kind shown in FIG.

  In order to make it easier to follow the beam path of the projection beam, the path of the principal ray CR at the outer field point of the off-axis object field OF is indicated by a bold line in FIGS. For the purposes of this application, the term “chief ray” (also known as the chief ray) refers to the outermost field point (farthest away from the optical axis) of the effectively used object field OF to the center of the entrance pupil. Represents a light beam extending. Due to the rotational symmetry of this system, the chief rays can be selected from corresponding field points in the meridian plane, as shown in the figure for illustrative purposes. In projection objectives that are essentially telecentric on the object side, the chief ray is launched from the object plane at a parallel or very small angle to the optical axis. The imaging process is further characterized by the trajectory of the peripheral rays. As used herein, a “marginal ray” is a ray that extends from an axial object field point (field point on the optical axis) to the edge of the aperture stop. This marginal ray cannot contribute to image formation due to vignetting when an off-axis effective object field is used. The imaging process is further characterized by a “marginal ray” trajectory. As used herein, a “marginal ray” is a ray that extends from an off-axis object field point (field point at a distance from the optical axis) to the edge of the aperture stop. The term “upper marginal ray” means a marginal ray that increases in distance in the propagation direction with respect to the optical axis, ie, escapes from the optical axis near the object plane. On the other hand, the term “lower marginal ray” means a marginal ray that decreases in distance to the optical axis in the propagation direction, ie, approaches the optical axis near the object plane. The chief and marginal rays and the marginal rays are selected to characterize the optical properties of the projection objective (see also the description with respect to FIG. 4). Such angles included between the selected light beam and the optical axis are represented by “principal light beam angle”, “peripheral light beam angle”, and the like. Such a radial distance between the selected light beam and the optical axis is represented by “principal ray height”, “peripheral ray height”, and the like.

  The projection objective 100 includes five optical element groups aligned along a straight (unfolded) common optical axis OA, that is, a first lens group LG1 having a positive refractive power immediately after the object plane, A mirror group MG having a positive refractive power as a whole immediately after the lens group, a second lens group LG2 having a positive refractive power immediately after the mirror group, and a negative refractive power immediately after the second lens group. It can be subdivided into the third lens group LG3 and the fourth lens group LG4 having positive refractive power immediately after the third lens group. The lens groups LG1 to LG4 are purely refractive, and the mirror group MG is purely reflective (only the reflecting surface).

The first lens group LG1 (also referred to as the front lens group) is designed to image the telecentric entrance pupil of the projection objective into the first pupil surface P1 with a strong positive refractive power, thereby It functions in the form of a Fourier lens group that performs one Fourier transform. This Fourier transform causes a relatively large chief ray angle CRA _P1 on the first pupil surface P1 of the order of 28 °. As a result, the pupil diameter at the first pupil surface is relatively small.

The emitted light that emerges from the first pupil surface P1 is incident on the first mirror M1 that faces the object side and has an aspherical concave mirror surface, and is intermediate at a certain distance downstream from the first mirror M1. An image IMI1 is formed. The emitted light is then reflected on a second mirror M2, designed as an aspheric concave mirror, and reflected at a tilt angle towards a third mirror M3 having a reflective surface containing the optical axis OA. The concave mirror surface of the third mirror is positioned in the second pupil surface P2, where the chief ray intersects the optical axis. A very large chief ray angle CRA _PM ≈42 ° is created at the second pupil surface, resulting in a small second pupil (Lagrange invariant). Radiation reflected from the third mirror M3 (pupil mirror PM) with a large principal ray angle is captured by reflection on the aspheric image side concave mirror surface of the fourth mirror M4, It has a refractive power designed to focus the emitted light beam toward the second intermediate image IMI2 at a distance just downstream from the fourth mirror M4.

  The optical design can be modified so that the mirror surface has a spherical shape instead of an aspheric shape. For example, the second and fourth mirrors M2 and M4 can be achieved as spherical mirrors. Furthermore, the second mirror M2 and the fourth mirror M4 are configured as separate mirrors having different surface features (surface shapes) and / or the first and third mirrors M1 and M3 have different surface features. It can be achieved as a separate mirror. In this case, at least one of the individual mirrors can be achieved as a spherical mirror instead of an aspherical mirror.

The chief ray height in the object plane OS (also referred to as the object height) is considerably smaller than the chief ray height in the second mirror M2 immediately upstream of the pupil mirror M3, and similarly, the fourth ray immediately downstream of the pupil mirror. Obviously, it is substantially less than the corresponding chief ray height in the mirror M4. In a preferred embodiment, the ratio between the chief ray height CRH _O at the object plane and the chief ray height CRH _M at the mirrors immediately upstream and downstream of the pupil surface is substantially greater than 1, for example 2 Greater than or greater than 2.5. In the embodiment of FIG. 1, this ratio is approximately 2.7 for both mirrors M2 and M4.

  The emitted light enters the mirror group at the mirror group entrance MGI close to the first pupil P1, and exits the mirror group at the mirror group exit positioned near the second intermediate image, that is, near the field surface. . A mirror group plane MGP aligned perpendicular to the optical axis and positioned between the first mirror vertex and the second mirror vertex of the mirror group causes the beam to exit the mirror group at the mirror group exit. Receive 5 passes before. Therefore, four reflections can be obtained in a space compact in the axial direction defined between the mirror group entrance and the mirror group exit.

  The second intermediate image IMI2 magnified with respect to the effective object field OF is a purely refractive objective unit (rear lens) including the second lens group LG2, the third lens group LG3, and the fourth lens group LG4. The image is also formed on the image plane IS. The narrowed portion CON of the projection beam indicated by the local minimum value of the beam diameter is formed in the negative lens region in the third lens group LG3. The second lens group LG2 has a positive refractive power, and essentially functions as a field lens group that forms an image near the exit pupil of the fourth mirror M4 closer to the reflective imaging mirror group. This allows subsequent lenses to be designed with a relatively small diameter that is optically free at short overall axial lengths. The third lens group LG3 has a negative refractive power, thereby forming a narrowed portion or “neck” of the beam diameter. By providing this negative lens group, the numerical aperture after the second intermediate image IMI2 can be increased. Despite the required minimum diameter of the aperture of the system at a small numerical aperture of the second intermediate image IMI2, the third lens group LG3 is the fourth lens between the subsequent third lens group and the third pupil surface P3. Together with a part of the lens group LG4, an inverse telecentric system having a compact axial length is formed.

  In an alternative description, the optical element of the projection objective 100 comprises a lens of the first lens group LG1 for imaging a mask pattern provided in the object field region in the first intermediate image IMI1, and the first A first imaging objective section including a mirror M1, a second imaging objective section including a pupil mirror PM for imaging a first intermediate image in a second intermediate image IMI2, and a second A third imaging objective part for forming an intermediate image of the first image in the image plane IS. The first objective part is catadioptric (having six lenses in LG1 and one concave mirror M1), and the second objective part is purely reflective (reflective). Imaging), the third objective part formed by the concave mirrors M2, M3 and M4 and formed by LG2, LG3 and LG4 is purely refractive. A first catadioptric objective having a magnification ratio (| β | = 2.1) defines the size of the first intermediate image IMI1 and cooperates with the second mirror M2 to project on the pupil mirror PM. Define the correction status of the beam. The absolute value of the chief ray height in the mirror M2 just upstream of the pupil mirror and in the M4 just downstream of the pupil mirror is significantly larger than the chief ray height in the object plane, which is an advantageous condition for small pupil size in the pupil mirror Is another expression. The large chief ray angle that is responsible for the small pupil size in the pupil mirror PM is captured by the fourth mirror M4 and the beam that converges towards the second intermediate image and image side refractive lens groups LG2, LG3, and LG4. It is formed. This part (rear lens group) is optimized to control imaging aberrations and achieve a large image-side numerical aperture NA = 0.93.

  A purely reflective (reflective imaging) mirror group MG can provide a strong overcorrection of Petzval sum that offsets the opposing effects of the positive refractive power of the upstream and downstream lenses of this mirror group. For this purpose, the mirror group MG includes a first concave mirror M1 disposed on the side opposite to the object field OF of the optical axis, a second concave mirror M2 disposed on the same side of the optical axis, and a pupil mirror. It comprises a third concave mirror M3 arranged on the optical axis so as to function as PM1, and a fourth concave mirror M4 arranged on the object field side. The mirror group entrance MGE is formed between the facing edges of mirrors M2 and M4 that are geometrically close to the first pupil surface P1 on the object side of the mirror group. The mirror group entrance MGE can be formed by a hole or hole in the common substrate of the mirrors M2 and M4. The mirror group exit MGO is located on the opposite side of the first mirror M1 next to the edge of the pupil mirror M3 outside the optical axis OA. As will be described in more detail with respect to FIG. 3, the pupil mirror M3 and the first mirror M1 can be formed on a common substrate to form a mirror pair, but separate substrates are possible as well. (See FIG. 2).

  Projection objective 100 is designed as a dry objective having an image-side numerical aperture NA = 0.93 at an operating wavelength λ = 193 nm. The size of the rectangular effective object field OF is 26 mm × 5.5 mm. Image field radius (radius) y '= 18 mm. The specifications are summarized in Table 1. The leftmost column lists the numbers of refractive surfaces, reflective surfaces, or separately designated surfaces, the second column lists the radius of curvature [mm] of the surface, and the third column , The distance d [mm] between that surface and the next surface, listing a parameter called the “thickness” of the optical element, the fourth column was used to fabricate the optical element The material is listed, and the fifth column lists the refractive index of the material. The sixth column lists the optically available transparent radius [mm] of the optical component. A radius of curvature r = 0 in the table means a flat surface (having an infinite radius).

Some surfaces in Table 1 are aspheric surfaces. Table 1A lists the relevant data for these aspheric surfaces, from which the sagittal or rise height p (h) of these surface features is It can be calculated as a function.
p (h) = [((1 / r) h ² ) / (1 + SQRT (1− (1 + K) (1 / r) ² h ² ))] + C1 · h ⁴ + C2 · h ⁶ +. . .
Here, the reciprocal value (1 / r) of the radius of curvature is the curvature at the surface apex of the surface in question, and h is the distance from the optical axis of a point on this surface. Thus, the sagittal or rise height p (h) represents the distance measured along the z-direction, i.e., along the optical axis, from the vertex of the surface in question. Constants K, C1, C2, etc. are listed in Table 1A.

  The projection objective 100 of FIG. 1 is an example of a catadioptric linear arrangement system that is optimized taking into account at least two conflicting requirements. First, the advantages of the linear arrangement structure (such as a mechanically stable mounting technique without a folding mirror) can be obtained with a large image-side numerical aperture while keeping the design object field of view moderately small. In this connection, vignetting control is an important issue. Second, a pupil mirror is provided, and dynamic or static control of imaging characteristics is possible by deforming the surface shape of the pupil mirror surface. Since the pupil mirror is arranged on the optical axis, it forms an obstacle to the projection beam, which makes vignetting control difficult. Third, it has been found that target control of imaging aberrations requires careful control of the correction status of the projection beam at the pupil mirror if it is desired to have an essentially constant effect across the field of view. Yes. The solution exemplarily represented by projection objective 100 complies with all these requirements (see also FIGS. 3 and 4).

Since the projection beam should be guided through the mirror in the linear arrangement design without vignetting, it is desirable to keep the area that affects vignetting as small as possible. In this example, this corresponds to the requirement to keep the projection beam pupil, i.e. the cross section of the projection beam at the pupil surface, as small as possible in the mirror region of the mirror group. With the Lagrangian invariant, this requirement translates to achieving a chief ray angle greater than normal at the pupil position in or near the mirror group. The large positive refractive power of the first lens group LG1 (Fourier lens group or front lens group) that bends the chief ray CR significantly towards the optical axis achieves a small pupil in the first pupil surface P1. A role is achieved, which allows for the extension of the reflective area of the small size mirror group entrance MGI and the mirrors M2 and M4 very close to the optical axis. In conjunction with the positive refractive power of the concave mirrors M1 and M2, the chief ray angle at the pupil mirror PM further increases to CRA _PM ≈42 ° and is small at the second pupil surface P2 on which the pupil mirror PM is placed. A beam diameter of the size results. Since the size of the reflection area RA3 of the pupil mirror PM (= M3) can be kept small, the vignetting control between the fourth mirror M4 and the image plane is facilitated, and the use of the first mirror M1. The reflection area RA1 and the use reflection area RA3 in the pupil mirror M3 can be separated. In other words, the light receiving areas of the projection beam do not overlap in reflection on the mirrors M1 and M3. This is one prerequisite for using the pupil mirror PM as a dynamically adjustable manipulator to dynamically influence the imaging characteristics of the projection objective.

  Furthermore, an emphasis is placed on the correction status of the projection beam at the pupil mirror PM, ie at the second pupil surface P2. Optimal conditions make it possible to obtain a constant field correction of imaging aberrations if partial apertures corresponding to different field points of the object field have the same size and shape and overlap completely within the pupil surface About that. If this condition is met, for example, a local change in the reflection characteristics of the pupil mirror by deforming the mirror surface will have a similar effect on all beam bundles originating from different field points. Creates a constant field of view effect in the image plane. On the other hand, if the partial apertures of different field points do not overlap within the pupil surface, local changes in the reflection characteristics of the pupil mirror will affect the beam bundles originating from different field points differently, thereby A change in the correction effect is created over time.

FIG. 4 illustrates these conditions with selected rays representing the ray bundle resulting from the field point FPO on the optical axis OA and the ray bundle resulting from the off-axis field point FP1, ie, the principal ray CR, the upper marginal ray RRU. , And the lower peripheral ray RRL. In the ideal case described above (in the pupil position, the partial apertures of the beam bundles of all the field points are completely overlapped), the principal ray CR must intersect the optical axis OA at the position of the reflecting surface of the pupil mirror PM. Don't be. Deviation from this ideal condition is expressed in the outermost field point FP1 of the principal ray CR entering the optical system, by a parameter CRH _i representing the ray height (radial distance from the optical axis) in the pupil mirror PM The This lateral offset must be small compared to the quantity D _O representing twice the peripheral ray height of the peripheral ray MR. Furthermore, the meridian diameter of the image in the pupil mirror PM of the entrance pupil of the objective with respect to the field point FP1 (represented by the parameter D _i ) should ideally coincide with the diameter D _O , in other words the ratio D _i / D _O must be equal to or close to 1. In the embodiment of FIG. 1, | CRH _i | / D value of _O = 0.03 and D _i / D _O = 0.991 is obtained. Substantially the same conditions apply for the sagittal section. In general, if the manipulation of the pupil mirror surface shape is to have an essentially constant effect on the image field correction status for all field points, the condition | CRH _i | / D _O < 0.1 and 0.9 <D _i / D _o <1.1 must be followed.

  FIG. 3 shows an enlarged detail view of the mirror group of FIG. 1 further illustrating the state present around the pupil mirror PM. From a structural point of view, the first mirror M1 and the third mirror M3 are a pair of concave mirrors formed on a common substrate. This substrate has a thick and mechanically rigid portion that provides a concave surface that holds the reflective layer that forms the first mirror M1. Integrated with the rigid part RP is a relatively thin and flexible part FP that holds the reflective coating for the pupil mirror PM. A concave portion is formed on the back side of the flexible portion FP in the mirror substrate. Several actuators (represented by arrows) of pupil mirror manipulator PMM are disposed in the recesses and are operatively coupled to the back side of flexible portion FP. The actuator is controlled by a pupil mirror control unit PMCU which can be an integral part of the central control unit of the projection exposure apparatus. The pupil mirror control unit is connected to receive a signal representative of the desired deformation of the pupil mirror surface. The pupil mirror manipulator and corresponding control unit can be designed essentially as disclosed in Applicant's US patent application US2004 / 0144915A1. The corresponding disclosure is incorporated herein by reference. Instead of the manipulators described above, any suitable structure of pupil mirror manipulators can be used, for example, manipulators using piezoelectric elements, actuators that respond to fluid pressure changes, and / or electromechanical actuators such as magnetic actuators. These actuators can be used to deform a continuous (unbroken) pupil mirror surface as described. The pupil mirror manipulator can also include one or more heating or cooling elements that produce a local temperature change of the mirror, causing the desired deformation of the pupil mirror surface. For this purpose, resistance heaters or Peltier elements can be used. The pupil mirror can also be designed as a multi-mirror array having a plurality of single micromirrors that are movable relative to each other in response to corresponding drive signals. Suitable multimirror arrays are disclosed, for example, in US 2006/0039669. The pupil mirror can be designed according to the principles disclosed in the applicant's international application published as WO2003 / 093903, the disclosure of which is also incorporated herein by reference.

  From an optical point of view, it is important to note that the utilization reflection area RA1 (indicated by a thick line) on the first mirror M1 does not overlap with the corresponding reflection area RA3 on the pupil mirror M3. Thereby, it is possible to change the shape of the pupil mirror surface without affecting the reflection generated in the first mirror M1. This design is also optimal with respect to the reflection area on the concave mirror and the caustic conditions in the projection beam in the first lens of the second lens group LG2 just downstream of the lens, especially the mirror group exit after the second intermediate image. It becomes. This is achieved by providing an intermediate image positioned at a relatively large distance away from the optical surface of the first lens of the mirror and lens group LG2 and essentially corrected for astigmatism and coma. The Avoiding caustic conditions in refractive or reflective optical surfaces helps to avoid significant maxima in the emitted light intensity and facilitates selective aberration control. Further, surface quality specifications can be relaxed by avoiding caustic on the surface.

  The embodiment of FIG. 1 can be modified to increase the option of manipulating imaging quality on a relatively short time scale. For example, the projection objective can include at least one further mirror having a mirror surface that can be manipulated using an associated manipulator operatively connected to the mirror surface. The pupil mirror that can be manipulated is typically placed at a position where the peripheral ray height MRH exceeds the chief ray height CRH, while still another mirror is optically closer to the field surface. Specifically, positioning at a position optically close to the field surface where the light height ratio MRH / CRH between the peripheral light height and the chief light height is less than 1 or even less than 0.5 Can do. An adaptive mirror (a mirror with a mirror surface that can be changed by a manipulator) positioned optically close to the field surface can be used to correct field dependent aberrations. In the modification of the first embodiment, the first mirror M1 of the mirror group MG positioned optically close to the adjacent field surface (intermediate image IMI1) is similar in structure and operation to the pupil mirror manipulator described above. It can be designed as an adaptive mirror by providing a field mirror manipulator that can. Since both the pupil mirror M3 and the field mirror M1 can be formed on the same substrate, the actuator design in the field mirror manipulator and the pupil mirror manipulator can be interlocked with each other to facilitate the structure. Alternatively or additionally, at least one of the second mirror M2 and the fourth mirror M4, both in optical proximity to the field surface, has a mirror surface that can be modified or changed using a manipulator. Can be designed as a mirror. Since both mirrors M2 and M4 can be formed on the same substrate, a common actuator mechanism can be used in this case.

FIG. 5 shows a straight line of a second embodiment having the general layout described in relation to FIGS. 1 and 2 with respect to the order and type of optical elements (lenses, mirrors) and the trajectory of the projection beam through the system. A placement projection objective 500 is shown. Please refer to the corresponding description. Elements and groups of elements having similar characteristics to those in the conventional embodiment are indicated with the same reference identification. The specifications for this design are summarized in Tables 5 and 5A.
Projection objective 500 has a high refractive index immersion liquid 1 between the exit surface of the projection objective and the image plane IS, for example λ = 193 nm with an image side numerical aperture NA = 1.2 when used with pure water. Designed as an immersion objective in This design is optimized for a rectangular effective image field with a field size of 26 × 5.5 mm ² that can be imaged without vignetting.

Similar to the embodiment of FIG. 1, the catadioptric first objective part including the first mirror M1 of the mirror group MG produces a first intermediate image IMI1 located in the inter-mirror space of the mirror group MG. . The second, third, and fourth mirrors M1 to M4 of the mirror group MG form a second reflective imaging subsystem that forms a second intermediate image IMI2 from the first intermediate image. The lens groups LG2, LG3, and LG4 form a third refractive objective unit that re-images the second intermediate image IMI2 on the image plane IS with a reduction scale (magnification ratio β = about −0.125). To do. The maximum lens diameter in the image-side bulge seen near the aperture stop AS positioned near the third pupil surface P3 between the constriction CON and the image plane IS is equal to that of the lower NA system of FIG. It is clear that it has increased in comparison. Nevertheless, the optically used diameter D _PM of the pupil mirror PM (mirror M3) remains relatively small, and the projection beam that has passed through the mirror can be guided without vignetting. The small pupil mirror size is the strong positive refractive power of the first lens group LG1 (which serves as the Fourier lens group forming the first pupil in P1), and the subsequent positive refraction of the mirrors M1 and M2. This is made possible by the force, and it becomes possible to obtain a chief ray angle CRA _PM ≈45 ° in the pupil mirror. In other words, the chief ray angle at the pupil mirror is further increased by the increased NA, allowing the size of the pupil mirror to be kept small according to the Lagrange invariant.
As described with reference to FIG. 3, a pupil mirror manipulator PMM is provided to deform the reflecting surface of the pupil mirror as necessary.

FIG. 6 shows that when used with a high refractive index immersion fluid having a refractive index n = 1.65, λ = 193 nm with an image side numerical aperture NA = 1.55 in a 26 × 5.5 mm ² annular field. FIG. 3 shows a third embodiment projection objective 600 designed for immersion lithography in FIG. The final optical element closest to the image plane IS is a plano-convex lens PCL made of LuAG (Lutetium Aluminum Garnet) having a refractive index = 2.14 at λ = 193 nm. The immersion liquid is cyclohexane with n _l = 1.65. This specification is provided in Tables 6 and 6A. This example shows that an extremely high numerical aperture can be obtained in a linear arrangement system having the pupil mirror PM (mirror M3) on the optical axis. The aperture stop AS near the third pupil surface P3 on the image side is positioned in the region of the beam that converges strongly between the maximum beam diameter region in the fourth lens group LG4 and the image plane IS. Although the image-side numerical aperture is significantly increased compared to the embodiment of FIG. 1, the size of the pupil mirror PM is due in part to the large chief ray angle CRA _PM ≈36 ° at the second pupil surface P2. And stay medium. In addition, vignetting control is facilitated by using an arcuate effective object field OF (annular field of view).

In all the embodiments described above, a catadioptric linear projection objective having a compact group of mirrors MG in the axial direction that provides four reflections is used, and the third mirror is a pupil mirror (operated as required). Can be arranged at the pupil position. At least two reflections at the concave mirror optically upstream of the pupil mirror are considered favorable for obtaining a large chief ray angle CRA _PM at the position of the pupil mirror, thereby reducing the small size pupil and the small size pupil. A mirror is possible. Furthermore, the small size pupil mirror allows a high aperture projection beam at a reasonably large effective object field size within a reasonably small design object field to be guided without vignetting through a compact group of mirrors.
FIGS. 8 and 9 show another embodiment of providing at least one pupil mirror that has four reflections in a compact mirror group MG and can be used as a dynamically controllable correction element for imaging aberrations. Figure 2 shows a catadioptric linear arrangement projection objective.

In the embodiment of FIGS. 1 to 6 described above, the mirror group entrance MGI is positioned near the pupil surface (first pupil surface) P1, but the mirror group exit MGO is in a region separated from the optical axis OA. Positioned optically close to the second intermediate image IMI2. The pupil mirror is provided at the third reflection in the mirror group.
In the embodiment of FIG. 7, the mirror group MG is arranged such that the mirror group entrance MGI is positioned optically close to the object plane OS outside the optical axis OA, ie optically close to the field surface. There is no lens or lens group between the object plane and the mirror group entrance MGI, but one or more lenses can be provided here. The first mirror M1, which is a convex surface, forms a first optical element and contributes to converging the emitted light toward the second mirror M2, which is a pupil mirror PM positioned on the optical axis OA. The third mirror M3 converges the emitted light so as to form a first intermediate image IMI1 positioned inside the catadioptric cavity of the mirror group. The catadioptric subsystem including the fourth mirror M4 directs the emitted light beam through the mirror group exit MGO positioned at the second pupil surface P2. The second intermediate image IMI2 is formed outside the mirror group and between the positive lens groups (indicated by arrows with arrows pointing outward). A subsequent refractive third objective part re-images the second intermediate image on the image plane. In this embodiment, the mirrors in the mirror group are basically used in reverse order when compared to the conventional embodiment. This design requires that the effective object field be located sufficiently away from the optical axis, which has a tendency to increase the diameter of the design object field and thereby vignetting moderately sized object fields It becomes more difficult to project with a high numerical aperture.

  In the embodiment schematically illustrated in FIG. 8, both the mirror group entrance MGI and the mirror group exit MGO are optically close to the field surface (ie optically away from the pupil surface) and positioned outside the optical axis OA. Is done. A refractive element arranged upstream of the mirror group entrance focuses the emitted light directly towards the first mirror M1, which is the first pupil mirror PM1 of the system. The second pupil is formed at the position of the fourth mirror M4 which is the second pupil mirror PM2 of the system after the reflection of the second mirror M2 and the third mirror M3. The first intermediate image IMI1 is formed between the second reflection and the third reflection, and the second intermediate image IMI2 is formed downstream of the fourth reflection. Positioned inside the space formed by the curved surface of the mirror. The second intermediate image IMI2 is re-imaged on the image plane by the subsequent refractive lens group. In order to dynamically compensate for imaging errors in the system, either the first mirror M1 or the fourth mirror M4, or both the first and fourth mirrors, shape the pupil mirror surface. It can be designed as an adaptive mirror that allows it to be manipulated. For a given object height, it is difficult to obtain a large chief ray angle in the first pupil mirror PM1 and on the second pupil mirror PM2, whereby the pupil mirror size is significantly increased with increasing image side numerical aperture. Will increase. This effect has a tendency to limit the ability to carry large geometric luminous flux (etendue) without vignetting. Also, in order to capture the divergent beam exiting the mirror group exit MGO, a relatively large lens is required immediately downstream of the mirror group. Preferably, this type of system can be used with a relatively large reduction ratio, eg 8: 1 instead of 4: 1, compared to a system with a small reduction ratio such as 4: 1. This is because the numerical aperture on the object side and the object field height can be reduced.

  FIG. 9 schematically shows a microlithographic projection exposure system in the form of a wafer scanner WS provided for producing large scale integrated semiconductor components in immersion scan mode using immersion lithography. The projection exposure system includes an excimer laser L having an operating wavelength of 193 nm as a light source. Other operating wavelengths are possible, for example 157 nm or 248 nm. The downstream illumination system ILL generates a large illumination field that is well-defined and uniformly illuminated at its exit surface ES, which is arranged off-axis with respect to the optical axis of the projection objective PO. And meet the telecentric requirements of the downstream catadioptric projection objective PO. The illumination system ILL has a device for selecting an illumination mode, in this example conventional on-axis illumination with varying degrees of interference and off-axis illumination, in particular annular illumination (in the pupil surface of the illumination system). With an annular illumination area) and dipole or quadrupole illumination.

Downstream of the illumination system, a mask is placed in the exit surface ES of the illumination system coinciding with the object plane OS of the projection objective PO, and in scanning operation, an optical axis OA (common to the illumination system and the projection objective). That is, a device RS (reticle stage) for holding and operating the mask M is arranged in such a manner that it can move in this plane in the scanning direction (Y direction) perpendicular to the Z direction). .
The size and shape of the illumination field IF provided by the illumination system specifies the size and shape of the effective object field OF of the projection objective that is actually used to project the pattern image on the mask into the image plane of the projection objective. To do. The slit-shaped illumination field IF has a height A parallel to the scanning direction, a width B> A perpendicular to the scanning direction, and is rectangular (as shown in the inset) or arcuate. (Annular field of view).

  The reduction projection objective PO is telecentric on the object side and image side and is designed to image an image of the pattern provided by the mask on a wafer W coated with a photoresist layer at a reduction scale of 4: 1. . Other reduction scales are possible, for example 5: 1 or 8: 1. The wafer W, which serves as a photosensitive substrate, is arranged in such a way that the flat substrate surface SS with the photoresist layer essentially coincides with the flat image plane IS of the projection objective. The wafer is held by a device WS (wafer stage) including a scanner driving device so that the wafer moves in parallel with the mask M in synchronization with the mask M. The device WS also includes a manipulator for moving the wafer in both the Z direction parallel to the optical axis OA and the X and Y directions perpendicular to the optical axis. A tilting device having at least one tilt axis extending perpendicular to the optical axis is integrated.

  A device WS (wafer stage) provided to hold the wafer W is configured for use in immersion lithography. The device WS includes a receptacle device RD, which can be moved by a scanner driver, whose bottom has a flat recess for receiving the wafer W. The periphery forms a flat, upwardly open liquid tight receptacle for the liquid immersion medium IM that can be introduced into the receptacle and drained from the receptacle using a device not shown. Given a correctly set working distance between the exit of the objective and the wafer surface, the height of the edge allows the filled immersion medium to completely cover the surface SS of the wafer W, and the projection It is dimensioned in such a way that the exit end region of the objective instrument PO can be immersed in the immersion liquid.

The projection objective PO has a plano-convex lens PCL as the final optical element closest to the image plane IS, and the flat exit surface of this lens is the last optical surface of the projection objective PO. During operation of the projection exposure system, the exit surface of the final optical element is completely immersed in the immersion liquid IM and is wetted by the immersion liquid IM. In an exemplary case, ultrapure water having a refractive index n ₁ ≈1.437 (193 nm) is used as the immersion liquid.

  In order to monitor the temperature of the immersion liquid IM during operation of the projection exposure system, a temperature sensor SENS is provided. For this purpose, a sensing element that is responsive to temperature changes is placed near the exit surface of the projection objective PO and monitors the temperature of the immersion layer that is subjected to transverse irradiation during exposure. The temperature sensor of an exposure system including a pupil mirror control unit PMCU provided to control the reflecting surface shape of the pupil mirror of the projection objective PO using a pupil mirror manipulator PMM (for example, compare with FIG. 3). Connected to the central control unit. The pupil mirror control unit PMCU includes a digital storage including a look-up table for converting the temperature signal supplied by the temperature sensor SENS into a value relating to the refractive index of the immersion liquid IM in the immersion layer. The temperature of the immersion layer may be due to absorption of the radiation intensity of the projection beam (temperature increase) or due to the inflow of new immersion liquid into the space between the exit surface of the projection objective and the wafer. (Temperature rise or fall) Since there is a possibility of changing during the exposure, the refractive index of the immersion layer may vary. These variations can cause a constant field of view contribution to spherical aberration that affects image formation on the wafer. To compensate for the effect of the immersion layer on spherical aberration, these variations in the optical properties of the image-side telecentric projection system adjust the pupil mirror's reflection shape so that a corresponding amount of spherical aberration is introduced by the pupil mirror. To be compensated for. A stable immersion lithography process is obtained using such a control loop.

  The pupil mirror control unit PMCU is configured to receive a signal indicating an illumination environment used for exposure from the illumination system ILL, and a control routine that allows the pupil mirror surface to be adjusted according to the selected illumination environment. including. For example, if the mask pattern to be projected on the wafer consists essentially of parallel lines extending in one direction, bipolar setting DIP (see left inset) to increase resolution and depth of focus Can be used. For this purpose, the tunable optical elements in the illumination system are adjusted and two locally concentrated strong diametrically opposite positions outside the optical axis OA in the pupil surface PS of the illumination system ILL. An intensity distribution is obtained which is characterized by an illumination area IR of light intensity and a light intensity which is little or not on the optical axis. A similar non-uniform intensity distribution is obtained in the pupil surface of the projection objective that is optically conjugate to the pupil surface of the illumination system. As a result, the lenses at or near the first and third pupil surfaces P1, P3 of the projection objective described above are each characterized by two “hot zones” that are diametrically opposed outside the optical axis. Can be exposed to a spatially non-uniform radiation load, which can cause lens heating induced by local absorption, causing unique lens deformation and refractive index changes, These cause a unique wavefront deformation characterized by essentially double radial symmetry with respect to the optical axis. Appropriate manipulation of the pupil mirror surface is utilized to compensate for these effects by providing an appropriate deformation of the pupil mirror surface with double radial symmetry in the correct orientation with respect to the optical axis.

Illumination environment is, for example, conventional illumination (rotational symmetry around the optical axis) or quadrupole illumination (quadruradial symmetry around the optical axis, right hand side inset QUAD with four off-axis illumination regions IR. The pupil mirror control unit supplies a corresponding signal to the pupil mirror manipulator, and the surface shape of the pupil mirror is changed accordingly.
Illumination systems that can optionally provide the off-axis pole illumination modes described above are described, for example, in US Pat. No. 6,252,647 B1, or the applicant's patent application US 2006 / 005026A1, the disclosure of these documents. Are incorporated herein by reference. Adaptation of the pupil mirror configuration to the illumination environment can be used not only in immersion systems such as those exemplified above, but also in dry systems, i.e. systems using dry objectives with NA <1. .
In another embodiment (not shown), the control signal to the pupil mirror manipulator provided to deform the reflecting surface of the pupil mirror during operation is an empirical value of a control parameter stored in the pupil mirror control unit or Derived from the calculated value. In these embodiments, direct or indirect measurement of the imaging properties of the projection system is not necessary.

A further catadioptric projection objective having a concave pupil mirror and a control system for controlling the shape of the reflecting surface of the pupil mirror will now be described with reference to FIGS.
FIG. 10 shows a catadioptric projection objective 1000 designed for a nominal UV operating wavelength λ = 193 nm. The layout of the projection objective with respect to the number, shape and position of lenses and other optical elements is taken from the prior art projection objective shown in FIG. 4 and described as a second embodiment in European patent EP1069448B1 It is. The disclosure of this reference is incorporated herein by reference. This projection objective is adapted for “dry lithography” in which the space between the exit surface and the image plane of the projection objective is filled with a gas. An image-side numerical aperture NA = 0.75 is obtained at a reduction ratio of 6: 1 (| β | = 1/6) in an arcuate off-axis image field. Other embodiments may have different magnification ratios, eg, | β | = 1/5, or | β | = 1/4, or | β | = 1 (unit magnification).

  The projection objective 1000 creates a flat image plane IS (image plane) from a pattern image from a mask (reticle) placed in a flat object plane OS (object plane) while creating exactly one real intermediate image IMI. Configured to project into. The first catadioptric objective part OP1 is designed to image a pattern from the object plane into the intermediate image IMI. The second purely refractive objective part OP2 forms an intermediate image directly in the image plane (i.e. without further intermediate image). Two mutually conjugate pupil surfaces P1 and P2 are formed at positions where CR intersects with the optical axis OA. The first pupil surface P1 is formed in the first objective instrument part, whereas the second pupil surface P2 is formed in the second objective instrument part OP2. The first objective part OP1 has only a moderate reduction effect and the main contribution to the overall reduction is provided by the refractive second objective part OP2. All the optical elements are aligned with a single straight optical axis OA, and the parallel orientation of the object plane OS and the image plane IS can be set. The exit pupil of the projection objective is substantially circular. The first concave mirror M1 is positioned very close to the first pupil surface P1, thereby forming a pupil mirror PM.

  The first objective instrument section OP1 includes a positive lens group LG1 formed by two positive meniscus lenses, a negative lens group LG2 formed by a single negative biconcave lens, and immediately downstream of the negative lens group LG2. The first concave mirror M1 having a reflecting surface facing the object plane, and the second concave mirror M2 having a concave surface facing the first concave mirror and the image plane. The second objective part OP2 includes a positive lens group LG3 formed by a single positive lens, a negative lens group LG4 formed by a single negative biconcave lens, and a second pupil surface P2. A positive lens group LG5 including five positive lenses and two negative lenses is provided between the image planes. The variable aperture stop AS that enables the use image-side numerical aperture NA to be adjusted is positioned on the second pupil surface between the fourth lens group and the fifth lens group.

  The emitted light from the object plane is converged toward the first concave mirror M1 by the positive first lens group LG1, and further reflected toward the second concave mirror M2 by the first concave mirror. The second concave mirror M2 converges the emitted light to form a first intermediate image. The emitted light that is guided toward and reflected from the first concave mirror M1 passes through the negative lens group LG2 twice in the opposite direction. Both the reflecting surface of the first concave mirror M1 and the negative lens group LG2 are such that the cross section of the emitted light beam is slightly deviated from the circular shape, and the peripheral ray height is at least 4 times the principal ray height or even at least 5 Positioned optically very close to the second pupil surface P2, which is twice as large. The combination of the concave mirror M1 disposed on or very close to the pupil surface and the negative lens group LG2 that is coaxial with the concave mirror M1 and through which the radiated light passes twice is the “Schupmann achromat” method. Acts to help correct chromatic aberration, especially axial chromatic aberration. The second objective instrument part OP2 re-images the intermediate image IMI to form a final image in the image plane IS.

  The pupil mirror manipulator PMM attached to the back side of the pupil mirror PM (M1) is configured to change the shape of the aspherical reflecting surface of the pupil mirror using an appropriate actuator (not shown). The general design of the pupil mirror manipulator can be the same as or different from that described in connection with FIG. The actuator is controlled by a pupil mirror control unit PMCU which can be an integral part of the central control unit of the projection exposure apparatus. The pupil mirror control unit is configured to generate a control signal toward an actuator of a pupil mirror manipulator that adjusts the surface shape of the pupil mirror to obtain a desired target shape.

The pupil mirror control unit PMCU is connected to the first sensor SENS1 and the second sensor SENS2. The first sensor SENS1 is an integral part of a measurement system, such as an interferometer, which allows to measure the imaging quality of an active projection objective and provide a signal indicative of the measured value representing the optical performance. To do. For example, the first sensor SENS1 can be configured to detect wavefront aberration of a wavefront incident on the image plane. An example of a wavefront measurement system suitable for this purpose is given, for example, in US 2002 / 0001088A1, the disclosure content of which is incorporated herein by reference. The second sensor SENS2 is configured to derive a signal indicating the current status of the aperture stop, thereby making it possible, for example, to derive the current image-side numerical aperture NA used for processing. Alternatively or additionally, the second sensor is arranged to derive a signal indicative of the intensity or intensity distribution in the pupil surface or a signal indicative of the wavefront characteristics in the pupil surface and / or field surface of the projection objective. be able to.
Similar control circuitry can be provided in embodiments having more than one intermediate image, such as two intermediate images, and / or in a folding system.

  FIG. 11 shows a catadioptric projection objective 1100. This projection objective is designed for a nominal UV operating wavelength λ = 193 nm. The layout of the projection objective with respect to the number, shape and position of lenses and other optical elements is shown in FIG. 19 and described in the fifth embodiment (Tables 9 and 10) in the international patent application WO2004 / 019128A2. Taken from a technical projection objective. The disclosure of each of these references is incorporated herein by reference. An image-side numerical aperture NA = 1.25 is obtained at a reduction ratio of 4: 1 within a rectangular off-axis image field of size 26 mm × 4 mm.

The projection objective 1100 generates two actual intermediate images IMI1 and IMI2 in a precise manner, and converts a pattern image on a reticle arranged in a flat object plane OS (object plane) into a flat image plane IS (image plane). For example, it is designed to project at a reduced scale of 4: 1. The rectangular effective object field OF and image field IF are off-axis, ie completely outside the optical axis OA. The first refractive objective instrument part OP1 is designed to image a pattern in the object plane into the first intermediate image IMI1. The second catadioptric (refractive / reflective) objective instrument part OP2 forms the first intermediate image IMI1 on the second intermediate image IMI2 at a magnification close to 1: (− 1). The third refractive objective instrument part OP3 forms the second intermediate image IMI2 on the image plane IS with a strong reduction ratio.
Three mutually conjugate pupil surfaces P1, P2, and P3 are formed at positions where the principal ray CR intersects the optical axis. The first pupil surface P1 is formed in the first objective part between the object plane and the first intermediate image, and the second pupil surface P2 is between the first intermediate image and the second intermediate image. And a third pupil surface P3 is formed in the third objective part between the second intermediate image and the image plane IS.

  The second objective part OP2 includes a single concave mirror CM on the second pupil surface P2, thereby forming a pupil mirror PM. The first flat folding mirror FM1 is optically close to the first intermediate image IMI1 and reflects 45 ° with respect to the optical axis OA so as to reflect the radiation coming from the object plane in the direction of the concave mirror CM. Arranged at an angle. The second folding mirror FM2 having a flat mirror surface aligned at right angles to the flat mirror surface of the first folding mirror has the radiated light coming from the concave mirror CM (pupil mirror PM) with respect to the object plane. Reflects in the direction of the parallel image plane. Each of the folding mirrors FM1 and FM2 is positioned optically close to the intermediate image.

  In this projection objective, a negative group is formed immediately before the concave mirror CM, and the emitted light travels from the first folding mirror FM1 toward the concave mirror along the path, and further from the concave mirror toward the second folding mirror FM2. And 27 lenses including two negative meniscus lenses that pass twice. A concave mirror located at or near the pupil surface and in a double pass region where the emitted light in front of the reflective side of the concave mirror passes through the negative group at least twice in the opposite direction The combination with a negative group comprising at least one negative lens is sometimes referred to as “Schupmann achromat”. This group contributes significantly to the correction of chromatic aberration, in particular axial chromatic aberration.

  The pupil mirror manipulator PMM is provided on the back side of the concave mirror CM that is freely accessible. The pupil mirror manipulator acts on the back side of the flexible mirror substrate so that the reflecting surface shape can be continuously changed according to the control signal supplied by the pupil mirror control unit PMCU connected to the pupil mirror manipulator PMM. Including several actuators (represented by arrows). The first sensor SENS1 is operatively connected to the flexible mirror substrate and enables the deformation state of the reflecting surface to be sensed two-dimensionally with high spatial resolution. The pupil mirror control unit is connected to the sensor SENS1 and receives a feedback signal representing the actual state of deformation of the pupil mirror surface. The pupil mirror manipulator and corresponding control unit can be designed essentially as described above in connection with FIG. The first sensor SENS1 may be an electromechanical sensor that derives an electrical signal from the mechanical state of the pupil mirror. Alternatively or additionally, a sensor system for monitoring the deformation state of the concave mirror can be designed according to the principle outlined in US 6,784,977 B2. The disclosure of each of this document is incorporated herein by reference.

  In general, the mirror arrangement formed by pupil mirrors and pupil mirror manipulators configured to deform or change the shape of the reflective pupil mirror surface can be configured as described above or in various other ways. . Examples are shown in US 5,986,795 or in the applicant's patent application US 2006 / 0018045A1. The disclosure of each of these documents is incorporated herein by reference.

The control of the pupil mirror manipulator to generate the desired deformation of the reflecting surface of the pupil mirror can be configured in various ways independent of the design type (folding or linear arrangement) and the number of intermediate images.
In some embodiments, the control system that controls the pupil mirror manipulator receives at least one input signal indicative of at least one state of the projection objective or another part of the projection exposure apparatus, and in response to the input signal A control circuit configured to output to the pupil mirror manipulator a control signal representing an adjustment of the surface shape of the pupil mirror to accommodate the imaging characteristics of the projection objective. The control circuit can be operated in an open loop control manner.

  For example, an input signal indicative of an illumination environment (eg, dipole illumination or quadrupole illumination) can be received from the illumination system and processed in a control circuit, and the expected non-uniform lens heating of the illumination is received. Deform the pupil mirror's reflective surface to the actuator of the pupil mirror manipulator to obtain a surface deformation with double or quadruple rotational symmetry, respectively, so that at least a portion is compensated by unequal deformation of the pupil mirror surface A control signal is generated. Alternatively or additionally, other input signals, for example, an input signal representing a pattern type (eg, a line pattern, a hole pattern, and / or a pattern having lines in different directions), an input signal representing a numerical aperture NA, And / or an input signal representing the exposure time can be generated and processed.

  Even higher optical performance stability and better response to disturbances can be obtained in embodiments that incorporate closed-loop control of projection objective performance. Unlike simple open loop control, closed loop control introduces feedback in the control circuit. In some embodiments, the control circuit includes at least one first sensor configured to detect a surface shape of the reflective surface of the pupil mirror, or a characteristic of the projection objective that correlates to the surface shape. Including one feedback circuit, which is connected to the pupil mirror control unit to provide a feedback signal, and the pupil mirror control unit optionally modifies the control signal controlling the pupil mirror manipulator in response to the feedback signal Configured as follows. For example, the optical performance of a wavefront measuring device or projection objective can be measured to generate a signal indicative of the level of aberrations present in the wavefront incident on the image plane and / or in the pupil surface. Another measurement system can be used. The level of aberration can be, for example, one or more monochromatic aberrations such as spherical aberration, coma, astigmatism, field curvature, and distortion, and / or axial and lateral chromatic aberration, and the color of monochromatic aberration. It can be characterized by one or more aberrations including one or more chromatic aberrations including changes. If the aberration exceeds a pre-determined threshold, the pupil mirror control unit will cause the level of critical aberration to be reduced below the threshold given by the end-user-specified specification typically for certain processes. A control signal for adjusting the surface shape of the pupil mirror can be generated. The apparatus for wavefront detection described in the applicant's patent application US2002 / 0011088A1 can be used in conjunction with a closed loop control that optimizes the surface shape of the pupil mirror.

The surface shape of the pupil mirror, or the characteristics of the projection objective that directly correlates with the surface shape, can be monitored constantly or intermittently to derive a feedback signal.
The at least one input signal described above that is processed in an open-loop control circuit or a closed-loop control circuit can be derived from parameters that can be derived from measurements on the projection objective, i.e. the parameters are directly in the system. Can be detected. It is also possible to derive one or more input signals from the simulation model, the simulation model being a projection objective or part thereof so that significant control parameters and signals can be derived from the simulation model. Or the entire projection exposure apparatus is reproduced with sufficient accuracy. In this case, pupil mirror control may include model-based control (MBC) aspects. For this purpose, the pupil mirror control unit can comprise a model data memory for storing model data representing model parameters of a projection objective and / or a simulation model of a projection exposure apparatus including the projection objective, or Can be connected. The control system can derive at least one input signal for the control circuit from the model data stored in the model data memory. The projection objective can include one or more second sensors for detecting an actual status parameter for the projection objective and derives an actual observable parameter corresponding to the model parameter of the simulation model. To do.

  For example, in the feedback control system shown in FIG. 11, the pupil mirror control unit PMCU stores a model data memory MDM that stores model data representing model parameters of a simulation model of a projection exposure apparatus including the projection objective 1100 and / or the projection objective. including. The model data memory MDM can be incorporated in an external device accessible by the pupil mirror control unit over the data network. In the embodiment described above, the model data memory MDM is temperature data representing the temperature of one or more components, temperature distribution data representing the spatial temperature distribution in one or more components, 1 Axial position of one or more components, position data representing at least one of eccentricity or tilt of one or more components, shape data representing the shape of the reflecting surface of the pupil mirror, aperture Aperture data representing the aperture conditions (using NA), setting data representing the illumination environment, radiant power data representing the power of the radiation source, one or more in the image field or pupil surface of the projection objective Immersion data representing at least one characteristic of the immersion medium, mask or other patterning means, including aberration data representing the spatial distribution of aberrations of the liquid, data indicating the presence or absence of the immersion medium Thus one of the data such as pattern data representing information on the type of provided patterns or than can be stored at most. The simulation model is calibrated at regular intervals using measurement parameters corresponding to the model parameters for which the data is stored to maintain a close correlation between the simulation model and the actual system in operation can do.

  Incorporating model-based control functions into the control phase for adaptive pupil mirrors, especially when some of the issues to be addressed by adaptive pupil mirrors, such as lens heating effects, are at certain times. Considering the fact that the dynamic effect varies with a constant, it may be advantageous. Furthermore, there is generally no simple relationship between the distribution of radiant energy in an optical system and the corresponding effect on optical performance. When a closed loop control circuit is used, the actual optical performance of the projection objective is compared over a predetermined period with a defined theoretical or desired value, typically given by user specifications. If a deviation occurs between the actual value and the desired value, this control circuit is effective to reduce the deviation by an appropriate operation that can include, for example, an operation on the pupil mirror. In general, such closed loop control is implemented to react to observation errors and eliminate or minimize these errors. In general, the integration aspect of model-based control can be used to implement predictive control of an optical system, so that at least some future changes expected in the system for which the simulation model is designed Can be taken into account. Thereby, the future action control can be obtained.

  A lithographic projection apparatus comprising a measurement system for measuring the time variation of the aberration of the projection system and a predictive control system for predicting the time variation of the aberration of the projection system based on the model parameters is disclosed, for example, in US 2006/0114437 A1. It is disclosed. To the extent that the concept can be modified to be used to control an adaptive pupil mirror, the disclosure content of this document is incorporated herein by reference.

  Constant or intermittent calibration of the simulation model can be performed based on signals derived from actual measurements of the characteristics of the modeled system. The property to be detected or identified in a physical system (eg, the entire projection objective or exposure apparatus) is the temperature of one or more components, the spatial temperature on one or more components Distribution, axial position of one or more components, eccentricity or tilt of one or more components, reflecting mirror shape of pupil mirror, aperture stop condition (using NA), The illumination environment, the power of the radiation source, the image field of the projection objective and / or the spatial distribution of one or more aberrations in the exit pupil, the wavefront in the pupil surface such as the exit pupil, the exit pupil, etc. One or more of the pattern information representing the intensity or spatial distribution of the intensity in the pupil surface, the type of pattern provided by the mask or another patterning means may be included. For example, the pattern information may read related pattern identification data from the mask and / or from data stored in a commutative memory component associated with each mask and / or from data stored in a permanent memory component. Can be derived from

  Various configurations of the control system for controlling the deformable mirror can be used in the claimed embodiments of the invention and other projection exposure systems, for example, without depending on the folding geometry and the number of intermediate images. . For example, projection objectives with more than two intermediate images are possible. A projection objective having three intermediate images can be designed essentially according to the teachings of the applicant's international patent application WO2005 / 040890A, the disclosure of which is incorporated herein by reference. Yes.

The above description of the preferred embodiment has been given by way of example. From the disclosure provided, those skilled in the art will not only understand the present invention and its attendant advantages, but will also find various changes and modifications to the disclosed structures and methods. Accordingly, it is intended to embrace all such changes and modifications that fall within the spirit and scope of the present invention as defined by the appended claims and equivalents thereof.
The entire contents of the claims are hereby incorporated by reference.

(Table 1)

(Table 1 continued)

(Table 1A)

(Table 2)

(Table 2 continued)

(Table 2A)

(Table 3)

(Table 3 continued)

(Table 3A)

(Table 4)

(Continued in Table 4)

(Table 4A)

(Table 5)

(Continued in Table 5)

(Table 5A)

(Table 6)

(Continued in Table 6)

(Table 6A)

100 catadioptric projection objective IF image field IMI1 first intermediate image IS image plane LG1 first objective part P1 first pupil surface

Claims (4)

A light source that generates main radiation,
An illumination system that forms the main radiation and generates illumination radiation incident on a mask carrying a pattern;
A projection objective for projecting an image of the pattern onto a radiation sensitive substrate, the projection objective comprising a pupil mirror and a pupil mirror manipulator;
A control system including a pupil mirror control unit configured to control the pupil mirror manipulator;
Including
The projection objective is a catadioptric projection objective for imaging a pattern from an object field located in the object plane of the projection objective onto an image field located in the image plane of the projection objective;
A first objective portion configured to image a pattern from the object plane into an intermediate image and having a first pupil surface;
A second objective part configured to form the intermediate image in an image plane and having a second pupil surface optically conjugate with the first pupil surface;
Including
The pupil mirror has a reflective pupil mirror surface positioned at or near one of the first and second pupil surfaces;
The pupil mirror manipulator is operatively connected to a pupil mirror and configured to change a shape of the pupil mirror surface;
The illumination system may include cod also at least one polar illumination environment having at least one of around double and quadruple the radial symmetry of the optical axis, is and configured pole to the illumination environment can be selected ,
The pupil mirror control unit is configured to receive a signal indicative of the polar illumination environment used for exposure from the illumination system, the control system having substantially double or quadruple radial symmetry. Configured to adjust the pupil mirror surface in response to a selected polar illumination environment ;
A projection exposure apparatus. The control system receives at least one input signal indicative of at least one status parameter of the projection objective or another part of the projection exposure apparatus, and determines the imaging characteristics of the projection objective in response to the input signal. A control circuit configured to output a control signal representing an adjustment of a surface shape of the pupil mirror for adaptation to the pupil mirror manipulator;
The projection exposure apparatus according to claim 1, wherein the measurement system capable of measuring the optical performance of the projection objective generates an input signal indicating an intensity distribution on the pupil surface. The pupil mirror control unit includes or is connected to a model data memory that stores model data representing model parameters of a simulation model of at least one of the projection objective and a projection exposure apparatus including the projection objective. And
The control system derives at least one input signal to the control circuit from the model data stored in the model data memory;
The projection exposure apparatus according to claim 1 or 2, wherein By an immersion liquid having a refractive index n _l substantially greater than 1 disposed between the exit surface of the projection objective and a substrate having a substrate surface disposed at the image plane of the projection objective Further comprising an immersion layer formed;
The pupil mirror control unit provides signals indicating imaging aberrations and conditions that affect imaging aberrations related to time-dependent variations in the optical properties of the projection system, including time-dependent changes in the optical properties of the immersion layer. Configured to receive,
The control system is configured to change the shape of the pupil mirror surface in response to the signal so that deformation of the shape of the pupil mirror surface at least partially compensates for the imaging aberration caused by the time dependent variation. Configured to change,
The projection exposure apparatus according to any one of claims 1 to 3, wherein the projection exposure apparatus is characterized.

Images